st: effect size in xtreg

[Eva Wölbert <evawoelbert@hotmail.com>]

Dear all,
 
 I am using a random effects model to analyze a 2x2 experimental design. Factor 1 is manipulated between subjects, and factor 2 is manipulated within subjects. We observe several trials per subject per condition, hence subject is the panelvariable.
 
 I am looking for a way to compute a standardized measure of effect size in the spirit of 
 
 Cohen's D = (mean1 - mean2)/SD 

 for the effect of interest, which is the interaction of the two factors. 

In particular, it's unclear to me how I have to compute the denominator in this case. Any references to other meaningful measures of effect size are also welcome. 

Thanks in advance
Eva 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/